DOI: 10.1148/radiol.2211001583
(Radiology. 2001;221:70-74.)
© RSNA, 2001
Fetal Cerebellum: US Appearance with Advancing Gestational Age1
Kazumasa Hashimoto, MD,
Takashi Shimizu, MD,
Koichiro Shimoya, MD,
Toru Kanzaki, MD,
James F. Clapp, MD and
Yuji Murata, MD
1 From the Department of Obstetrics and Gynecology, Case Western Reserve University at MetroHealth Medical Center, Cleveland, Ohio (K.H., J.F.C.); Department of Obstetrics and Gynecology, Faculty of Medicine, Osaka University, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan (K.H., K.S., T.K., Y.M.); and Shimizu Women’s Clinic, Hyogo, Japan (T.S.). Received September 26, 2000; revision requested November 3; final revision received April 30, 2001; accepted April 11. Address correspondence to K.H. (e-mail: hashi@gyne.med.osaka-u.ac.jp).
 |
ABSTRACT
|
|---|
PURPOSE: To evaluate the change in ultrasonographic (US) appearance of the fetal cerebellum with advancing gestation.
MATERIALS AND METHODS: A total of 291 normal fetuses of uncomplicated pregnancies were evaluated at gestational ages (GAs) between 15 and 41 weeks with a 3.75-MHz transabdominal curvilinear probe. After the transcerebellar view was obtained, the transverse cerebellar diameter (TCD) was measured and the images were stored. On hard-copy US images, cerebella were assigned three grades of appearance. These grades were analyzed in relation to GA and TCD. Inter- and intraobserver variations were assessed in 91 randomly selected cases.
RESULTS: Cerebella in 137 (47.1%), 71 (24.4%), and 83 (28.5%) of 291 subjects were classified as grade I (hypoechoic, "eyeglass" shape), grade II (intermediate echogenicity, "dumbbell" outline), and grade III (hyperechoic, "fan" shape), respectively. With advancing gestation, the dominant grade changed from I to III gradually and progressively. The median GA and TCD, respectively, were 22 weeks and 22 mm for grade I, 29 weeks and 35 mm for grade II, and 36 weeks and 46 mm for grade III. These differences were statistically significant (P < .001). The agreements within inter- and intraobserver estimations were 96% (87 of 91) and 95% (86 of 91), respectively.
CONCLUSION: A gradual change in US appearance of the fetal cerebellum is seen with advancing gestation.
Index terms: Brain, growth and development, 153.92 • Fetus, growth and development, 153.92 • Fetus, US, 153.1298
|
INTRODUCTION
|
|---|
Recently, the evaluation of the posterior fossa of the fetal cranium has been accepted as part of the routine obstetric ultrasonographic (US) examination (1). With this evaluation, structural anomalies in the central nervous system such as Arnold-Chiari malformations and Dandy-Walker malformations can be detected. In addition, the size of the cerebellum or transverse cerebellar diameter (TCD) is also important because it is a useful biometric parameter in estimating gestational age (GA) in the second trimester (2). Indeed, in some cases with dolichocephaly or brachycephaly, TCD may be a more reliable predictor than biparietal diameter, since the posterior fossa is not affected by external pressure, which may induce distortion of the fetal head (2). Because TCD seems unaffected by intrauterine growth restriction (IUGR) (3,4), measuring TCD is especially advantageous when IUGR is suspected or when GA is uncertain.
In contrast to the progressive increase in dimension, however, the US appearance of the fetal cerebellum in normal and abnormal pregnancies has not been well described. The human cerebellum changes its histologic morphology in utero and after birth (5,6). In rhesus monkeys, water content of the fetal cerebellum has been shown to decrease progressively (7). Theoretically, these processes can be monitored with serial US examination throughout pregnancy, and preliminary observations in our laboratory indicate that the human fetal cerebellum changes its appearance with advancing gestation. Therefore, the goal of the present study was to evaluate the change in the US appearance of the fetal cerebellum with advancing gestation.
|
MATERIALS AND METHODS
|
|---|
Subjects
Consecutive women with normal singleton pregnancy who were undergoing routine antenatal US screening between 15 and 41 weeks gestation were asked to participate in the study. A total of 291 cases were included after verbal informed consent was obtained from each patient. All subjects were native Japanese and were recruited from a low-risk obstetric clinic. The protocol was approved by the institutional review board at the local clinic. Mean maternal age (±SD) was 28.9 years ± 3.3 (range, 20–39 years). In all cases, gestational age was confirmed with US crown-rump length at 8–10 weeks gestation. The cases were excluded from the study if patients had medical or obstetric complications including diabetes mellitus, thyroid disease, hypertension, or preeclampsia or in cases with substance abuse. Subjects were also excluded if the fetus had structural anomalies, chromosomal abnormalities, or abnormal growth.
US Examination
All US measurements were performed by a single perinatologist (T.S.) using a real-time system with a 3.75-MHz curvilinear transducer (model SSA-340A; Toshiba, Tokyo, Japan). After the transcerebellar view was obtained, TCD was measured as described previously (2,8) and the images were printed and stored. Each hard-copy image was interpreted, and the appearance of the cerebellum was assigned to one of three grades by a primary perinatologist (T.S.) (observer 1) according to the criteria listed below. The accuracy and precision of the definitions and criteria for the cerebellar grades were confirmed by a second perinatologist (K.H.) (observer 2) over the first 100 cases. Six months after the initial evaluation by observer 1, 91 of the cases were randomly selected and independently analyzed by observer 2 twice 1 month apart to calculate inter- and intraobserver agreement.
Grading of the Cerebellum
Grade I criteria.—(a) Each cerebellar hemisphere is round and (b) the vermis has not developed well, which make the whole cerebellar appearance that of "a pair of eyeglasses" at US. (c) The hemispheres lack echogenicity. Thus, the cerebellum appears to be two fluid-containing cysts (Fig 1).
Grade II criteria.—(a) The vermis can be seen more prominently and appears as echogenic rectangular tissue connecting the two hemispheres, which changes the whole cerebellar appearance to that of a "dumbbell." (b) Each hemisphere is oval, and the central portion is more echogenic than the peduncles and other background structures but less echogenic than the circumferential margin of the hemispheres and the vermis, which give the internal portion of the hemispheres a ground-glass appearance (Fig 2).
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Grade II cerebellum. Transcerebellar US view at 29 weeks gestation. Cerebellar hemispheres are oval and echogenic; echogenicity is prominent at the circumferential margin of the hemispheres and the vermis. The internal portion of the hemispheres represents a ground-glass-like appearance. The vermis has developed, and the cerebellar shape has changed to a "dumbbell-like" outline, shown between the crosshairs.
|
|
Grade III criteria.—(a) The appearance of the hemispheres changes to that of a triangular or "fan-shaped" structure. (b) Tissues in the central portion of the hemispheres show similar echogenicity to that of the margin and the vermis. Consequently, the cerebellum looks more like solid tissue than cyst (Fig 3).
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Grade III cerebellum. Transcerebellar US view at 34 weeks gestation. Cerebellar hemispheres show "fan" shape between the crosshairs, and the whole cerebellum is homogeneously echogenic, indicating a solid appearance.
|
|
Statistical Analyses
The two assessments by observer 2 were used to calculate intraobserver agreement. The first assessment of observer 2 was compared with the assessment by observer 1 for calculation of interobserver agreement. The medians of GA and TCD among the three cerebellar grade groups were compared by using Kruskal-Wallis nonparametric analysis of variance. Subsequently, the Dunn multiple comparison analysis was used to identify possible significant differences between any two grades. The analyses were performed with use of computer software (GRAPHPAD INSTAT version 3.01 for Windows 95/NT; GraphPad Software, San Diego, Calif). Interobserver agreement and intraobserver agreement were evaluated by using weighted 
coefficient computer software (METSTAT version 2.01; MetroHealth Medical Center and Case Western Reserve University, Department of Pediatrics, Cleveland, Ohio). A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS
|
|---|
A total of 291 cases were graded by observer 1: 137 (47.1%) were assigned grade I; 71 (24.4%), grade II; and 83 (28.5%), grade III. The median GA and TCD were 22 weeks and 22 mm for grade I, 29 weeks and 35 mm for grade II, and 36 weeks and 46 mm for grade III. There were significant differences among the three cerebellar groups in terms of GA and TCD by using Kruskal-Wallis nonparametric analysis of variance (P < .001) and the Dunn multiple comparison analysis between any two groups (P < .001). The percentage of each cerebellar grade is presented with respect to GA and TCD (Figs 4, 5), which shows gradual and continuous shift from grade I to grade II to grade III. All 84 cases before 23 weeks showed grade I cerebellum, and all 48 cases after 36 weeks showed grade III. Between 24 and 35 weeks, 53 (33.3%), 71 (44.7%), and 35 (22.0%) of 159 cases were grades I, II, and III, respectively. The grade II cerebellum was distributed over a 12-week period, and all three grades were found in each GA group between 28 and 31 weeks. Similarly, when TCD was 23 mm or less (n = 80) or 44 mm or greater (n = 60), all the cases were grade I and III, respectively. Between 24 and 43 mm, 57 (37.7%), 71 (47.0%), and 23 (15.2%) of 151 cases were grades I, II, and III, respectively.
View larger version (69K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Proportion of cerebellar US grades in GA range. The dominant cerebellar grade shows gradual and continuous shift from grade I to grade II to grade III with advancing gestation. Dark gray bars = grade I, light gray bars = grade II, black bars = grade III.
|
|
View larger version (63K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Proportion of cerebellar US grades in transcerebellar diameter range. The dominant cerebellar grade shows gradual and continuous shift from grade I to grade II to grade III with development of the cerebellum. Dark gray bars = grade I, light gray bars = grade II, black bars = grade III.
|
|
Among 91 randomly selected cases, observer 1 assigned 47 cases as grade I, 34 cases as grade II, and 10 cases as grade III. The second observer’s first assignment changed only one case from grade I to II, two from grade II to I, and one from grade II to III. The interobserver agreement was highly significant at 96% (87 of 91) with a weighted coefficient of 0.934 (Table 1). The second observer’s sequential evaluation resulted in changing three cases from grade II to I, one from grade I to II, and one from grade II to III. Thus, the intraobserver agreement was also highly significant at 95% (86 of 91) with a weighted coefficient of 0.923 (Table 2).
|
DISCUSSION
|
|---|
Our study findings clearly suggest a gradual and steady change in US appearance of the fetal cerebellum with advancing gestation. The US changes include changes in both shape and echogenicity. The US appearance of the cerebellum seems to change progressively from an "eyeglass" (grade I), to a "dumbbell" (grade II), and finally to a "fan" shape (grade III) with advancing gestation. The echogenicity of the cerebellum appears to change from hypoechoic (cystic) to slightly echoic with marginal enhancement (ground glass), and to homogeneously hyperechoic (solid) near term. These changes may be a reflection of the histologic development of the fetal cerebellum during pregnancy.
The results of the present cross-sectional study demonstrated the progressive change in the ratio of each cerebellar grade with advancing GA (Fig 4) and TCD (Fig 5). The predominant cerebellar grade was grade I up to 27 weeks, grade II between 28 and 32 weeks, and grade III thereafter. The proportion of grade I cases seemed to constantly decrease between 23 and 32 weeks, whereas that of grade III cases increased between 27 and 36 weeks. From the results shown in Figures 4 and 5, in the population presented in this report the major point of transition of cerebellum from grade I to II may be at about 27–28 weeks or 31–32 mm in TCD and that from grade II to III may be at about 32–33 weeks or 41–42 mm in TCD, where the predominant grade changes. According to the results from the cross-sectional study, it is likely and strongly suggested that cerebellar grade in an individual fetus changes in the same manner, that is, from I to II and II to III, with advancing gestation or cerebellar development. However, this needs to be confirmed in a serial US study. It is also a matter of interest that there was a wide variety in cerebellar grade between 24 and 35 weeks gestation. The grade II cerebellum was distributed over a 12-week period, and all three grades were found in each GA group between 28 and 31 weeks. Although the absolute duration of a grade II cerebellum cannot be estimated from this study, assuming that all the cases were "normal" and "uncomplicated" and grade II is a "transitional" stage in development, there may be a difference for months in the timing of these US changes among individuals, which reflects the rate of anatomic development. To answer the above questions, further prospective serial studies are under way.
The speculation that the fetal cerebellum develops gradually but steadily is supported by a great deal of histologic data in both animals and humans. The degree of Purkinje cell differentiation has been found to be a useful marker of cerebellar maturation in various species such as mice (9,10), rats (11), and monkeys (12–14). In animal studies, it has been shown that Purkinje cell development passes through three stages: (a) the stage of the fusiform cell, (b) the stage of the multipolar stellate cell, and (c) the stage of orientation and of flattening of the dendritic tree. This stage transition proceeds in a gradual and continuous fashion. In human fetuses, similar developmental stage transition was observed by Zecevic and Rakic (6). They demonstrated that the cerebellar development of the human fetus is strictly associated with Purkinje cell differentiation. Using the Golgi method and analyzing the cerebellar cytologic changes with electron microscopy, they found that development of the human fetal cerebellum undergoes three stages. The first stage occupies 12–16 weeks gestation, the second stage occupies 16–28 weeks gestation, and the third stage extends throughout the remaining period of intrauterine life and early postnatal years. During the first stage, Purkinje cells have only a few processes at the apical and basal cell poles and are distributed in a layer, which consists of several rows in depth. In the second stage, Purkinje cells become gradually organized into a single row. The somas come to have randomly oriented dendritic processes and numerous somatic spines. Synapse formation becomes more prominent at 18–24 weeks. Between 24 and 28 weeks, the number of somatic spines reaches its peak. In the third stage, somatic spines disappear and cell body and the stems of primary dendrites become relatively smooth. Although the relationship between the histologic stages and our US grading is unknown, there may be a positive relationship between the Purkinje cell development and the US appearance of the cerebellum because the timing of the shift from stage 2 to 3 and that from grade I to II are similar.
Another possibility to explain the transition of US appearance in fetal cerebellum is the alteration of the water content within the cerebellar tissue. It has been demonstrated that water content of rhesus monkey cerebellar cortex keeps decreasing during pregnancy and after birth (7). In that report, water content of the cerebellar cortex was about 88% at 40–60 days gestation, which was decreased to 79% at term, and then further decreased to 77% at over 5 years of age or adult stage in rhesus monkeys. In human fetuses, similar changes may also take place, which would explain the progressive increase in echogenicity of the cerebellum.
It is well known that the development of the cerebellum is affected by several obstetric factors such as IUGR and chromosomal abnormalities. Studies using fetal sheep demonstrated that the area of Purkinje cell dendritic tree and the growth of granule cell dendrites were substantially reduced or restricted by IUGR (15). Malnutrition, hypoxia, and hypothyroidism have been thought to contribute to these changes in synaptogenesis in fetuses with IUGR. It is also shown that development and maturation of the central nervous system are restricted and disorganized in fetuses with trisomy 21 (16–19), which may explain, at least partially, mental retardation and predisposition to early development of neuronal degeneration comparable to those found in Alzheimer disease (20). In addition, reports have suggested that at US, the TCD is affected by trisomy 18 (21) and trisomy 21 (22), which probably represents cerebellar hypoplasia. Therefore, chromosomal abnormalities may cause developmental restriction not only in size but in tissue maturation, which might be detectable with US grading of the cerebellar hemispheres and vermis. The timing of the shifting of the cerebellar grading might be restricted in fetuses with abnormal karyotype.
Other situations that may affect fetal cerebellar development are maternal corticosteroid administration and diabetes. Antenatal corticosteroid treatment, especially repeated doses, may reduce the size and weight of the fetal brain including the cerebellum in the experimental animals and possibly humans (23–25). In diabetic pregnancy, tissue protein content and myelin content were increased in the guinea pig fetal cerebellum (26,27).
Our results demonstrate that evaluation of the rate of fetal cerebellar development may be feasible by using a conventional US technique serially. Thus, alterations in cerebellar development similar to those described in the preceding paragraphs may be detectable during pregnancy by using US. In addition, the wide individual variation in cerebellar appearance between 24 and 35 weeks gestation in normal pregnancies suggests that extrinsic factors may also influence the timing of the progressive change in cerebellar appearance. Again, serial studies will help to answer this question. In any case, the grading system described herein may be useful to understand the normal development of the human cerebellum and to evaluate abnormal growth and maturation in normal and complicated pregnancies.
|
FOOTNOTES
|
|---|
Abbreviations: GA = gestational age,
TCD = transverse cerebellar diameter,
IUGR = intrauterine growth restriction
Author contributions: Guarantors of integrity of entire study, K.H., Y.M.; study concepts, K.H., T.S., T.K.; study design, K.H., T.S., K.S.; literature research, K.H., T.K.; clinical studies, K.H., T.S.; data acquisition, T.S., K.H., K.S.; data analysis/interpretation, K.H., T.S., J.F.C., Y.M.; statistical analysis, K.H., T.K.; manuscript preparation, K.H., T.S., K.S.; manuscript definition of intellectual content and editing, K.H.; manuscript revision/review, T.K., J.C., Y.M.; manuscript final version approval, all authors.
|
REFERENCES
|
|---|
-
American College of Obstetricians and Gynecologists. ACOG technical bulletin: ultrasonography in pregnancy No. 187. Washington, DC: American College of Obstetricians and Gynecologists, 1993.
-
McLeary RD, Kuhns LR, Barr M, Jr. Ultrasonography of the fetal cerebellum. Radiology 1984; 151:439-442.[Abstract/Free Full Text]
-
Reece EA, Goldstein I, Pilu G, Hobbins JC. Fetal cerebellar growth unaffected by intrauterine growth retardation: a new parameter for prenatal diagnosis. Am J Obstet Gynecol 1987; 157:632-638.[Medline]
-
Lee W, Barton S, Comstock CH, Bajorek S, Batton D, Kirk JS. Transverse cerebellar diameter: a useful predictor of gestational age for fetuses with asymmetric growth retardation. Am J Obstet Gynecol 1991; 165:1044-1050.[Medline]
-
Rakic P, Sidman RL. Histogenesis of cortical layers in human cerebellum, particularly the lamina dissecans. J Comp Neurol 1970; 139:473-500.[CrossRef][Medline]
-
Zecevic N, Rakic P. Differentiation of Purkinje cells and their relationship to other components of developing cerebellar cortex in man. J Comp Neurol 1976; 167:27-47.[CrossRef][Medline]
-
Selzer ME, Myers RE, Holstein SB. Maturational changes in brain water and electrolytes in rhesus monkey with some implications for electrogenesis. Brain Res 1972; 45:193-204.[CrossRef][Medline]
-
Hill LM, Martin JG, Fries J, Hixson J. The role of the transcerebellar view in the detection of fetal central nervous system anomaly. Am J Obstet Gynecol 1991; 164:1220-1224.[Medline]
-
Rakic P, Sidman RL. Organization of cerebellar cortex secondary to deficit of granule cells in weaver mutant mice. J Comp Neurol 1973; 152:133-161.[CrossRef][Medline]
-
Rakic P, Sidman RL. Sequence of developmental abnormalities leading to granule cell deficit in cerebellar cortex of weaver mutant mice. J Comp Neurol 1973; 152:103-132.[CrossRef][Medline]
-
Altman J. Postnatal development of the cerebellar cortex in the rat. II. Phases in the maturation of Purkinje cells and of the molecular layer. J Comp Neurol 1972; 145:399-463.[CrossRef][Medline]
-
Kornguth SE, Anderson JW, Scott G. The development of synaptic contacts in the cerebellum of Macaca mulatta. J Comp Neurol 1968; 132:531-546.[CrossRef][Medline]
-
Rakic P. Neuron-glia relationship during granule cell migration in developing cerebellar cortex: a Golgi and electronmicroscopic study in Macacus rhesus. J Comp Neurol 1971; 141:283-312.[CrossRef][Medline]
-
Rakic P, Stensas LJ, Sayre E, Sidman RL. Computer-aided three-dimensional reconstruction and quantitative analysis of cells from serial electron microscopic montages of foetal monkey brain. Nature 1974; 250:31-34.[CrossRef][Medline]
-
Rees S, Harding R. The effects of intrauterine growth retardation on the development of the Purkinje cell dendritic tree in the cerebellar cortex of fetal sheep: a note on the ontogeny of the Purkinje cell. Int J Dev Neurosci 1988; 6:461-469.[CrossRef][Medline]
-
Wisniewski KE, Laure-Kamionowska M, Wisniewski HM. Evidence of arrest of neurogenesis and synaptogenesis in brains of patients with Down’s syndrome. N Engl J Med 1984; 311:1187-1188.[Medline]
-
Petit TL, LeBoutillier JC, Alfano DP, Becker LE. Synaptic development in the human fetus: a morphometric analysis of normal and Down’s syndrome neocortex. Exp Neurol 1984; 83:13-23.[CrossRef][Medline]
-
Schmidt-Sidor B, Wisniewski KE, Shepard TH, Sersen EA. Brain growth in Down syndrome subjects 15 to 22 weeks of gestational age and birth to 60 months. Clin Neuropathol 1990; 9:181-190.[Medline]
-
Golden JA, Hyman BT. Development of the superior temporal neocortex is anomalous in trisomy 21. J Neuropathol Exp Neurol 1994; 53:513-520.[Medline]
-
de la Monte SM, Hedley-Whyte ET. Small cerebral hemispheres in adults with Down’s syndrome: contributions of developmental arrest and lesions of Alzheimer’s disease. J Neuropathol Exp Neurol 1990; 49:509-520.[Medline]
-
Hill LM, Marchese S, Peterson C, Fries J. The effect of trisomy 18 on transverse cerebellar diameter. Am J Obstet Gynecol 1991; 165:72-75.[Medline]
-
Rotmensch S, Goldstein I, Liberati M, Shalev J, Ben-Rafael Z, Copel JA. Fetal transcerebellar diameter in Down syndrome. Obstet Gynecol 1997; 89:534-537.[Abstract]
-
Velazquez PN, Romano MC. Corticosterone therapy during gestation: effects on the development of rat cerebellum. Int J Dev Neurosci 1987; 5:189-194.[CrossRef][Medline]
-
Huang WL, Beazley LD, Quinlivan JA, Evans SF, Newnham JP, Dunlop SA. Effect of corticosteroids on brain growth in fetal sheep. Obstet Gynecol 1999; 94:213-218.[Abstract/Free Full Text]
-
French NP, Hagan R, Evans SF, Godfrey M, Newnham JP. Repeated antenatal corticosteroids: size at birth and subsequent development. Am J Obstet Gynecol 1999; 180:114-121.[CrossRef][Medline]
-
Saintonge J, Cote R. Fetal brain development in diabetic guinea pigs. Pediatr Res 1984; 18:650-653.[Medline]
-
Saintonge J, Cote R. Brain development in relation to fetal weight and maternal glucose tolerance during normal gestation. Brain Dev 1987; 9:26-32.[Medline]
TOP
ABSTRACT
INTRODUCTION
